Spotlight on Recipients of the Kevin and AJ KleinOsowski and Gary H. Glover Fellowships

Susmita Dey Manasi, Zamshed Chowdhury, and Omer Demirel are some of the recipients of the KleinOsowski, and Glover Fellowships. Here, we shine a spotlight on their research interests, the impact of their work, and how the fellowships have affected their University experience. 

Susmita Dey Manasi (KleinOsowski 2017 – 2018)

The Kevin and AJ KleinOsowski fellowship not only supported me financially, but has also been a great motivator for me as I entered the doctoral program in the electrical engineering program. The fellowship has allowed me the luxury of staying focused on my research.

Doctoral student Susmita Dey Manasi

My name is Susmita Dey Manasi and I am pursuing a doctoral degree in the Department of Electrical and Computer Engineering, under the supervision of Prof. Sachin Sapatnekar. I am from Bangladesh where I earned my bachelor of science degree from the Bangladesh University of Engineering and Technology. My research interest lies in the design and optimization of VLSI circuits and systems. I am also interested in neuromorphic computing, approximate computing, beyond-CMOS technologies, as well as device-circuit and architectural co-design for developing efficient VLSI systems.

I chose the University of Minnesota after rigorously exploring the research initiatives in my areas of interest, as offered by several universities. The state-of-the-art research facilities, and outstanding faculty profiles relevant to my interests and experience, are what motivated me to apply to this University.

Deep learning (DL) networks have drawn a lot of attention because of the promise they hold for brain-inspired intelligence in electronics. Currently, DL techniques have a wide range of applications such as object classification, speech recognition, face detection, web search, natural language processing, driverless car, and others. However, DL tasks are computationally expensive. To enable widespread deployment of DL techniques, especially on resource-limited embedded platforms, the development of energy efficient hardware systems for DL applications is critical. Currently, I am working on DL hardware systems to enable low energy computation for DL networks. As part of my research, I have developed an analytical model to estimate neural computation energy on a customized hardware platform and utilized the model to determine energy optimal computation scheduling for DL tasks on mobile platform. Currently, I am looking into various architectural alternatives of embedded DL processors to design optimal hardware set for various DL applications.

After earning my doctoral degree, I see myself as an academic conducting research at a world-class institution, or as a scientist in a leading research lab.

Zamshed Chowdhury (KleinOsowski 2016 – 2017)

Being awarded the Kevin and AJ KleinOsowski fellowship was incredibly helpful to me. It allowed me to concentrate on my research and studies, and succeed in an educational system that is significantly different from the one I am familiar with. The fellowship was also vital in my decision to attend the University, as it helped to defray the cost of attendance.

Doctoral student Zamshed Chowdhury

My name is Zamshed Chowdhury, and I earned my bachelor’s and master’s degrees in applied physics, electronics, and communication engineering from the University of Dhaka in Bangladesh. Currently, I am pursuing my doctoral degree under the guidance of Prof. Ulya Karpuzcu. My research interests include emerging memory technology, high speed and energy efficient computing, and application specific hardware accelerator design.

My decision to apply to the University of Minnesota was driven by the excellence of the research groups in my areas of research interest. And the award of the KleinOsowski fellowship helped me finalize my school and program choice.

My current research projects involve designing in-memory processing solutions using emerging memory technologies. The goal is to reduce overheads due to expensive memory transfer in processing large datasets, which will make the processing of such datasets faster and energy-efficient. Such a design will also potentially address the approaching limits of Moore’s law. I was engaged in teaching for a while, back in Bangladesh. I plan to go back to teaching after graduation.

Omer Demirel (Glover 2017 – 2018)

The Gary H. Glover fellowship has been instrumental in my decision to pursue my graduate education at the University of Minnesota as opposed to other schools. Knowing that the fellowship awards process is highly competitive, receiving the Glover fellowship was confidence inducing. Critically, the fellowship provided me with a more financially viable year.

Doctoral student Omer Demirel

My name is Omer Demirel. I am from Turkey and I am a second year doctoral student in electrical engineering working under the supervision of Prof. Mehmet Akçakaya. I earned my bachelor’s and master’s degrees in electrical engineering from Bilkent University, where I also started research on Magnetic Particle Imaging.

My research interests lie in the field of magnetic resonance imaging (MRI), specifically image reconstruction algorithms for accelerated imaging, for instance cardiac imaging. Currently I am working on new image reconstruction methods that can reduce scan times without deteriorating image quality. My goal is to accurately diagnose cardiac diseases using novel techniques. The opportunities for cross-disciplinary research that the University of Minnesota offers was what drove me to apply to the doctoral program here. The department and faculty collaboration with the University’s Center for Magnetic Resonance Research (CMRR) which provides expertise in imaging applications, is a unique learning environment and opportunity for me.

On graduating, I would like to pursue a career in academia; my goal is to become a professor and establish my own lab, ideally in MRI research.